Spaces:
Runtime error
Runtime error
File size: 3,635 Bytes
2478285 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 |
import gradio as gr
import os
os.system('cd monotonic_align && python setup.py build_ext --inplace && cd ..')
import logging
numba_logger = logging.getLogger('numba')
numba_logger.setLevel(logging.WARNING)
import librosa
import torch
import commons
import utils
from models import SynthesizerTrn
from text.symbols import symbols
from text import text_to_sequence
import numpy as np
import soundfile as sf
from preprocess_wave import FeatureInput
def resize2d(x, target_len):
source = np.array(x)
source[source<0.001] = np.nan
target = np.interp(np.arange(0, len(source)*target_len, len(source))/ target_len, np.arange(0, len(source)), source)
res = np.nan_to_num(target)
return res
def transcribe(path, length, transform):
featur_pit = featureInput.compute_f0(path)
featur_pit = featur_pit * 2**(transform/12)
featur_pit = resize2d(featur_pit, length)
coarse_pit = featureInput.coarse_f0(featur_pit)
return coarse_pit
def get_text(text, hps):
text_norm = text_to_sequence(text, hps.data.text_cleaners)
if hps.data.add_blank:
text_norm = commons.intersperse(text_norm, 0)
text_norm = torch.LongTensor(text_norm)
print(text_norm.shape)
return text_norm
hps_ms = utils.get_hparams_from_file("configs/vctk_base.json")
net_g_ms = SynthesizerTrn(
len(symbols),
hps_ms.data.filter_length // 2 + 1,
hps_ms.train.segment_size // hps_ms.data.hop_length,
n_speakers=hps_ms.data.n_speakers,
**hps_ms.model)
featureInput = FeatureInput(hps_ms.data.sampling_rate, hps_ms.data.hop_length)
hubert = torch.hub.load("bshall/hubert:main", "hubert_soft")
_ = utils.load_checkpoint("G_312000.pth", net_g_ms, None)
def vc_fn(input_audio,vc_transform):
if input_audio is None:
return "You need to upload an audio", None
sampling_rate, audio = input_audio
# print(audio.shape,sampling_rate)
duration = audio.shape[0] / sampling_rate
if duration > 45:
return "Error: Audio is too long", None
audio = (audio / np.iinfo(audio.dtype).max).astype(np.float32)
if len(audio.shape) > 1:
audio = librosa.to_mono(audio.transpose(1, 0))
if sampling_rate != 16000:
audio = librosa.resample(audio, orig_sr=sampling_rate, target_sr=16000)
source = torch.FloatTensor(audio).unsqueeze(0).unsqueeze(0)
print(source.shape)
with torch.inference_mode():
units = hubert.units(source)
soft = units.squeeze(0).numpy()
audio22050 = librosa.resample(audio, orig_sr=16000, target_sr=22050)
sf.write("temp.wav", audio22050, 22050)
pitch = transcribe("temp.wav", soft.shape[0], vc_transform)
pitch = torch.LongTensor(pitch).unsqueeze(0)
sid = torch.LongTensor([0])
stn_tst = torch.FloatTensor(soft)
with torch.no_grad():
x_tst = stn_tst.unsqueeze(0)
x_tst_lengths = torch.LongTensor([stn_tst.size(0)])
audio = net_g_ms.infer(x_tst, x_tst_lengths, pitch=pitch,sid=sid, noise_scale=0.4,
noise_scale_w=0.1, length_scale=1)[0][0, 0].data.float().numpy()
return "Success", (hps_ms.data.sampling_rate, audio)
app = gr.Blocks()
with app:
with gr.Tabs():
with gr.TabItem("Basic"):
vc_input3 = gr.Audio(label="Input Audio (30s limitation)")
vc_transform = gr.Number(label="transform",value=1.0)
vc_submit = gr.Button("Convert", variant="primary")
vc_output1 = gr.Textbox(label="Output Message")
vc_output2 = gr.Audio(label="Output Audio")
vc_submit.click(vc_fn, [ vc_input3,vc_transform], [vc_output1, vc_output2])
app.launch() |